Prior research in representation engineering has revealed that LLMs encode concepts within their representation spaces, predominantly centered around English. In this study, we extend this philosophy to a multilingual scenario, delving into multilingual human value concepts in LLMs. Through our comprehensive exploration covering 7 types of human values, 16 languages and 3 LLM series with distinct multilinguality, we empirically substantiate the existence of multilingual human values in LLMs. Further cross-lingual analysis on these concepts discloses 3 traits arising from language resource disparities: cross-lingual inconsistency, distorted linguistic relationships, and unidirectional cross-lingual transfer between high- and low-resource languages, all in terms of human value concepts. Additionally, we validate the feasibility of cross-lingual control over value alignment capabilities of LLMs, leveraging the dominant language as a source language. Drawing from our findings on multilingual value alignment, we prudently provide suggestions on the composition of multilingual data for LLMs pre-training: including a limited number of dominant languages for cross-lingual alignment transfer while avoiding their excessive prevalence, and keeping a balanced distribution of non-dominant languages. We aspire that our findings would contribute to enhancing the safety and utility of multilingual AI.
相關內容
With the enhancement in the field of generative artificial intelligence (AI), contextual question answering has become extremely relevant. Attributing model generations to the input source document is essential to ensure trustworthiness and reliability. We observe that when large language models (LLMs) are used for contextual question answering, the output answer often consists of text copied verbatim from the input prompt which is linked together with "glue text" generated by the LLM. Motivated by this, we propose that LLMs have an inherent awareness from where the text was copied, likely captured in the hidden states of the LLM. We introduce a novel method for attribution in contextual question answering, leveraging the hidden state representations of LLMs. Our approach bypasses the need for extensive model retraining and retrieval model overhead, offering granular attributions and preserving the quality of generated answers. Our experimental results demonstrate that our method performs on par or better than GPT-4 at identifying verbatim copied segments in LLM generations and in attributing these segments to their source. Importantly, our method shows robust performance across various LLM architectures, highlighting its broad applicability. Additionally, we present Verifiability-granular, an attribution dataset which has token level annotations for LLM generations in the contextual question answering setup.
Automated driving systems are an integral part of the automotive industry. Tools such as Robot Operating System and simulators support their development. However, in the end, the developers must test their algorithms on a real vehicle. To better observe the difference between reality and simulation--the reality gap--digital twin technology offers real-time communication between the real vehicle and its model. We present low fidelity digital twin generator and describe situations where automatic generation is preferable to high fidelity simulation. We validated our approach of generating a virtual environment with a vehicle model by replaying the data recorded from the real vehicle.
Blockchain technology is leading a revolutionary transformation across diverse industries, with effective governance being critical for the success and sustainability of blockchain projects. Community forums, pivotal in engaging decentralized autonomous organizations (DAOs), significantly impact blockchain governance decisions. Concurrently, Natural Language Processing (NLP), particularly sentiment analysis, provides powerful insights from textual data. While prior research has explored the potential of NLP tools in social media sentiment analysis, there is a gap in understanding the sentiment landscape of blockchain governance communities. The evolving discourse and sentiment dynamics on the forums of top DAOs remain largely unknown. This paper delves deep into the evolving discourse and sentiment dynamics on the public forums of leading DeFi projects: Aave, Uniswap, Curve DAO, Yearn.finance, Merit Circle, and Balancer, focusing primarily on discussions related to governance issues. Our study shows that participants in decentralized communities generally express positive sentiments during Discord discussions. Furthermore, there is a potential interaction between discussion intensity and sentiment dynamics; higher discussion volume may contribute to a more stable sentiment from code analysis. The insights gained from this study are valuable for decision-makers in blockchain governance, underscoring the pivotal role of sentiment analysis in interpreting community emotions and its evolving impact on the landscape of blockchain governance. This research significantly contributes to the interdisciplinary exploration of the intersection of blockchain and society, specifically emphasizing the decentralized blockchain governance ecosystem. We provide our data and code for replicability as open access on GitHub.
Many observational studies feature irregular longitudinal data, where the observation times are not common across individuals in the study. Further, the observation times may be related to the longitudinal outcome. In this setting, failing to account for the informative observation process may result in biased causal estimates. This can be coupled with other sources of bias, including non-randomized treatment assignments and informative censoring. This paper provides an overview of a flexible weighting method used to adjust for informative observation processes and non-randomized treatment assignments. We investigate the sensitivity of the flexible weighting method to violations of the noninformative censoring assumption, examine variable selection for the observation process weighting model, known as inverse intensity weighting, and look at the impacts of weight trimming for the flexible weighting model. We show that the flexible weighting method is sensitive to violations of the noninformative censoring assumption and show that a previously proposed extension fails under such violations. We also show that variables confounding the observation and outcome processes should always be included in the observation intensity model. Finally, we show that weight trimming should be applied in the flexible weighting model when the treatment assignment process is highly informative and driving the extreme weights. We conclude with an application of the methodology to a real data set to examine the impacts of household water sources on malaria diagnoses.
Text generation has become more accessible than ever, and the increasing interest in these systems, especially those using large language models, has spurred an increasing number of related publications. We provide a systematic literature review comprising 244 selected papers between 2017 and 2024. This review categorizes works in text generation into five main tasks: open-ended text generation, summarization, translation, paraphrasing, and question answering. For each task, we review their relevant characteristics, sub-tasks, and specific challenges (e.g., missing datasets for multi-document summarization, coherence in story generation, and complex reasoning for question answering). Additionally, we assess current approaches for evaluating text generation systems and ascertain problems with current metrics. Our investigation shows nine prominent challenges common to all tasks and sub-tasks in recent text generation publications: bias, reasoning, hallucinations, misuse, privacy, interpretability, transparency, datasets, and computing. We provide a detailed analysis of these challenges, their potential solutions, and which gaps still require further engagement from the community. This systematic literature review targets two main audiences: early career researchers in natural language processing looking for an overview of the field and promising research directions, as well as experienced researchers seeking a detailed view of tasks, evaluation methodologies, open challenges, and recent mitigation strategies.
Finding the root causes of anomalies in cloud computing systems quickly is crucial to ensure availability and efficiency since accurate root causes can guide engineers to take appropriate actions to address the anomalies and maintain customer satisfaction. However, it is difficult to investigate and identify the root causes based on large-scale and high-dimension monitoring data collected from complex cloud computing environments. Due to the inherently dynamic characteristics of cloud computing systems, the existing approaches in practice largely rely on manual analyses for flexibility and reliability, but massive unpredictable factors and high data complexity make the process time-consuming. Despite recent advances in automated detection and investigation approaches, the speed and quality of root cause analyses remain limited by the lack of expert involvement in these approaches. The limitations found in the current solutions motivate us to propose a visual analytics approach that facilitates the interactive investigation of the anomaly root causes in cloud computing systems. We identified three challenges, namely, a) modeling databases for the root cause investigation, b) inferring root causes from large-scale time series, and c) building comprehensible investigation results. In collaboration with domain experts, we addressed these challenges with RCInvestigator, a novel visual analytics system that establishes a tight collaboration between human and machine and assists experts in investigating the root causes of cloud computing system anomalies. We evaluated the effectiveness of RCInvestigator through two use cases based on real-world data and received positive feedback from experts.
In the evolving landscape of machine learning, a pivotal challenge lies in deciphering the internal representations harnessed by neural networks and Transformers. Building on recent progress toward comprehending how networks execute distinct target functions, our study embarks on an exploration of the underlying reasons behind networks adopting specific computational strategies. We direct our focus to the complex algebraic learning task of modular addition involving $k$ inputs. Our research presents a thorough analytical characterization of the features learned by stylized one-hidden layer neural networks and one-layer Transformers in addressing this task. A cornerstone of our theoretical framework is the elucidation of how the principle of margin maximization shapes the features adopted by one-hidden layer neural networks. Let $p$ denote the modulus, $D_p$ denote the dataset of modular arithmetic with $k$ inputs and $m$ denote the network width. We demonstrate that a neuron count of $ m \geq 2^{2k-2} \cdot (p-1) $, these networks attain a maximum $ L_{2,k+1} $-margin on the dataset $ D_p $. Furthermore, we establish that each hidden-layer neuron aligns with a specific Fourier spectrum, integral to solving modular addition problems. By correlating our findings with the empirical observations of similar studies, we contribute to a deeper comprehension of the intrinsic computational mechanisms of neural networks. Furthermore, we observe similar computational mechanisms in the attention matrix of the one-layer Transformer. This research stands as a significant stride in unraveling their operation complexities, particularly in the realm of complex algebraic tasks.
In pace with developments in the research field of artificial intelligence, knowledge graphs (KGs) have attracted a surge of interest from both academia and industry. As a representation of semantic relations between entities, KGs have proven to be particularly relevant for natural language processing (NLP), experiencing a rapid spread and wide adoption within recent years. Given the increasing amount of research work in this area, several KG-related approaches have been surveyed in the NLP research community. However, a comprehensive study that categorizes established topics and reviews the maturity of individual research streams remains absent to this day. Contributing to closing this gap, we systematically analyzed 507 papers from the literature on KGs in NLP. Our survey encompasses a multifaceted review of tasks, research types, and contributions. As a result, we present a structured overview of the research landscape, provide a taxonomy of tasks, summarize our findings, and highlight directions for future work.
The existence of representative datasets is a prerequisite of many successful artificial intelligence and machine learning models. However, the subsequent application of these models often involves scenarios that are inadequately represented in the data used for training. The reasons for this are manifold and range from time and cost constraints to ethical considerations. As a consequence, the reliable use of these models, especially in safety-critical applications, is a huge challenge. Leveraging additional, already existing sources of knowledge is key to overcome the limitations of purely data-driven approaches, and eventually to increase the generalization capability of these models. Furthermore, predictions that conform with knowledge are crucial for making trustworthy and safe decisions even in underrepresented scenarios. This work provides an overview of existing techniques and methods in the literature that combine data-based models with existing knowledge. The identified approaches are structured according to the categories integration, extraction and conformity. Special attention is given to applications in the field of autonomous driving.
A fundamental goal of scientific research is to learn about causal relationships. However, despite its critical role in the life and social sciences, causality has not had the same importance in Natural Language Processing (NLP), which has traditionally placed more emphasis on predictive tasks. This distinction is beginning to fade, with an emerging area of interdisciplinary research at the convergence of causal inference and language processing. Still, research on causality in NLP remains scattered across domains without unified definitions, benchmark datasets and clear articulations of the remaining challenges. In this survey, we consolidate research across academic areas and situate it in the broader NLP landscape. We introduce the statistical challenge of estimating causal effects, encompassing settings where text is used as an outcome, treatment, or as a means to address confounding. In addition, we explore potential uses of causal inference to improve the performance, robustness, fairness, and interpretability of NLP models. We thus provide a unified overview of causal inference for the computational linguistics community.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191